mgalkin/ultra_50g
Graph Machine Learning • 169k • Updated • 28 • 59
HomeWork
-
-
HiFi-UMI: Learning Deployable Manipulation Policies from High-Fidelity UMI Data Alone
Paper • 2607.25895 • Published • 156 -
Kimi K3: Open Frontier Intelligence
Paper • 2607.24653 • Published • 495 -
Co-Evolution in Agentic Systems: Toward Self-Directed Evolution Beyond Human Design
Paper • 2608.10299 • Published • 131 -
Beyond Pixels: From Video Priors to 4D Worlds
Paper • 2608.10744 • Published • 178 -
AdvFD: Boosting Visual Generation via Adversarial Fr'echet Distance Loss
Paper • 2608.11205 • Published • 27 -
PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives
Paper • 2608.13552 • Published • 44 -
Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus
Paper • 2608.12149 • Published • 30 -
Full-bandwidth transformer
Paper • 2608.08888 • Published • 21 -
Knowing When to Quit: Diagnosing and Training LLMs to Abort Futile Reasoning
Paper • 2607.29211 • Published • 13 -
Maglev: Sliding Recurrent Memory
Paper • 2608.02870 • Published • 14 -
Thought-Level Beam Search for Reasoning
Paper • 2608.08020 • Published • 14 -
CW-BASS v2: Saturation-Aware Pseudo-Label Selection for Semi-Supervised Segmentation under Foundation-Model Teachers
Paper • 2608.12773 • Published • 8 -
RibAssist 3D: Biplanar Rib-Fracture Detection, Addressing, and Selective 3D Localization from CT-Derived Projections
Paper • 2608.06914 • Published • 9 -
Large Discovery Models: Empirically-grounded Model-Based Open-Ended Search
Paper • 2608.15669 • Published • 46